By Robert D. Schneider
Thanks to the tremendous expansion of audio, video, images, XML documents, and other fresh data sources, modern applications must now contend with vastly larger amounts of information - much of which is semi-structured or even unstructured. This opens the door to new and inventive applications capable of extracting value from all of this data. Naturally, developers and database administrators are responsible for making this happen, yet in many cases they’re hampered by technologies that haven’t kept up.
In response to today’s realities, Sybase Adaptive Server Enterprise (ASE) has been updated to offer a collection of highly targeted features that make it possible to realize the benefits from these new and larger information sources. In this article, we’ll pay particular attention to illustrating how the data-driven application advantages offered by Sybase ASE help developers and database administrators produce solutions that deliver tangible business value.
The intended audience for this article includes executives, line-of-business owners, IT leaders, database administrators, and developers – in short, anyone with a desire to utilize data to boost productivity, profitability, and customer satisfaction.
What Developers Need to Produce Data-driven Applications
Regardless of their particular industry or domain, developers confront three sobering truths when creating software:
1. Operational data stores are now routinely measured in terabytes. In addition, annual information growth rates often exceed 100%.
2. Unstructured data is increasingly becoming an essential part of IT’s operations. These new information sources are often at the heart of the organization’s most strategic applications.
3. High performance remains essential. In fact, in a frantic attempt to cope with today’s processing workloads, enterprises are deploying technologies such as parallel hardware.
By supplying greatly simplified programmatic interfaces, well-architected capabilities for both structured and unstructured data, and enhanced SQL syntax, Sybase ASE helps developers and DBAs surmount the obstacles we just itemized. To help bring these advances to life, we’ll now explore three specific examples of ASE in action, including:
· Seamlessly supporting unstructured data
· Working with very large data volumes
· Delivering faster performance
For each advantage, we’ll describe how it was implemented along with how it furnishes business benefits.
Seamlessly Supporting Unstructured Data
As we declared earlier, today’s IT environments feature a dramatically broader array of information categories than even just a few years ago. Many of these new data classes contain unstructured data, in contrast with the row-and-column-oriented structured data that has been traditionally stored in tables within relational databases. Examples of these varied unstructured information types include:
· Document scans
· XML files
Commonly known as Large OBjects (LOB), each instance of unstructured data can be substantial, ranging in size from megabytes to gigabytes. Most applications must contend with thousands or even millions of LOBs.
ASE supplies a collection of LOB-oriented features that make it easier to create applications adept at working with unstructured data. These include:
· Plentiful LOB capacity. Each LOB instance can hold up to 2 GB of data, which is large enough to store just about any imaginable LOB.
· LOB locator. This identifier serves as a pointer to the LOB, which saves applications from needing to transmit the entire LOB itself. This results in reduced network traffic and faster performance.
· Client-side LOB interaction. All ASE LOB features are available from client-side JDBC and ODBC applications, which makes it possible to deliver complete, end-to-end LOB-friendly applications.
ASE’s unstructured data capabilities directly translate to business benefits: it’s now possible to deliver pioneering applications that can fully exploit these advanced unstructured data types, without forcing developers and DBAs to learn new tools and techniques. No ASE reconfiguration is necessary, either. Taken together, this represents a much more agile and cost-effective approach.
Working with Larger Information Volumes
An assortment of technological, business, and regulatory trends have joined forces to noticeably boost the amount of data collected and managed by most enterprises. Some of the most prominent factors include:
· New data sources. Mobile devices, RFID, scanners, and social media are just a few of the culprits that are flooding the enterprise with information.
· Longer data retention regulations. Driven by copious government and industry mandates, these standards are placing unprecedented data management burdens on organizations.
· Increased numbers of transactions. Even the most staid enterprises are processing more of these events.
· Business intelligence (BI) technology. Previously restricted to only the largest and most technologically advanced enterprises, BI is now employed by more users in more situations than ever before.
· Unstructured data. In the previous section, we portrayed the potential for each instance of unstructured data to be massive.
ASE presents developers and DBAs with techniques to help them make the most of all this data. These include:
· MERGE statement. This command makes it much easier and more efficient to load vast amounts of data into ASE via the ‘upsert’ capability: any new information will be inserted, while any existing rows will be updated. This is much faster and simpler than alternative approaches, and doesn’t require any developer or DBA intervention.
· Optional in-row LOB storage. In other database platforms, LOBs are stored on separate pages from structured data. This frequently leads to bloated storage and diminished performance. In contrast, ASE evaluates each LOB to determine its optimal storage: in some instances, the LOB will be kept on the same page as its structured siblings, while in other situations it will be stored on a separate page. This intelligence and flexibility results in faster access and fewer wasted storage assets.
Business intelligence and analytic applications are transforming the ways many enterprises conduct their operations. The end results of these new solutions include enhanced revenue opportunities, superior competitive positioning, and increased operational efficiencies. However, these new analytic applications are voracious information consumers, meaning that ASE’s support for expanded data volumes is a prerequisite before these benefits can be realized.
Delivering Faster Performance
There’s never been a better time to be a technology consumer: powerful, highly functional software is available everywhere from the desktop to the smartphone. Unfortunately, from IT’s perspective, users are also more demanding than ever before. They’ve been conditioned to expect near-instantaneous system response time, which is particularly challenging when you consider all of the new data sources, volumes, and categories that we’ve been discussing.
Rather than being compelled to rewrite, tweak, or otherwise adjust their applications to deliver better performance, developers are able to leverage key ASE throughput-related capabilities. Two of the most beneficial are:
· Statement caching. The ASE engine caches all SQL text. If the same query is run by the same user, the pre-existing query plan is used. In addition, ASE tracks and employs supplementary attributes aimed at lowering CPU usage and assisting the query optimizer. The end results are noticeably reduced system overhead and substantially better throughput.
· Reduced latency. ASE supplies an assortment of internal optimizations meant to condense execution time overhead. First, ASE utilizes superior metadata caching algorithms. Secondly, it employs global caching for dynamic SQL statements. This shared caching approach is better for both single user as well as multiuser access, and cursor handling is better, too. Finally, ASE imposes minimal overhead for query execution, especially in the context of SELECT/INSERT/UPDATE/DELETE statements.
Performance obstacles have blocked many businesses from completely realizing the benefits of unstructured data and larger information volumes. By addressing performance concerns, ASE lets you run the applications that are able to leverage all of these new varieties and quantities of information.
It should come as no surprise that IT will likely continue to be inundated with new classes of information and much larger quantities of data. And as always, developers will be expected to put all of this intelligence to work. Sybase ASE contains substantial developer-oriented capabilities that make creating and maintaining these new applications much easier, using familiar tools and techniques.
About the Author
Robert D. Schneider is a Silicon Valley-based technology consultant and author. He has provided database optimization, distributed computing, and other technical expertise to a wide variety of enterprises in the financial, technology, and government sectors. He has written six books and numerous articles on database technology and other complex topics such as cloud computing, Big Data, data analytics, and Service Oriented Architecture (SOA). He is a frequent organizer and presenter at technology industry events, worldwide.